soil & more international b.v. -...

Soil & More International of 45

Soil & More International B.V.

March 2010

Comprehensive Carbon Footprint Assessment

Dole Bananas


Title Comprehensive Carbon Footprint Assessment

Dole Bananas

Author Boki Luske

T: +31 188 007 95 02

[email protected]

Date March 2010

Copyright No part of this publication may be reproduced

in any form by print, photo print, microfilm or

any other means without permission of Soil &

More International.

Disclaimer Neither Soil & More International, nor its

partners, accepts any liability whatsoever for

any direct or consequential loss however

arising from any use of this document or its

contents or otherwise arising in connection

herewith.

Contact

address

Soil & More International

Transportweg 23

2742 RH Waddinxveen

The Netherlands

T: +31 088 007 95 00

F: +31 (0)180 890 420

W: www.soilandmore.com


Content

Content ........................................................................................................ 3

Summary ...................................................................................................... 4

Acronyms and Glossary ................................................................................ 5

1 Public Awareness and Market Developments ............................................ 7

2 General information ................................................................................ 10

2.1 Introduction ................................................................................. 10

2.2 Company profile ............................................................................ 10

2.3 Background information ................................................................. 10

2.4 Goals of a Carbon Footprint ............................................................ 11

2.5 Functional unit .............................................................................. 11

3 Methodology............................................................................................ 12

3.1 General methodology .................................................................... 12

3.2 System boundary and scopes ......................................................... 12

3.3 Data sources ................................................................................ 14

3.4 Allocation with co-production .......................................................... 15

3.5 Exclusions .................................................................................... 15

4 Greenhouse Gas Inventory ...................................................................... 16

4.1 Farming stage .............................................................................. 16

4.2 Packaging stage ............................................................................ 20

4.3 Terminal and port operations and overseas transport ........................ 22

4.4 Ripening stage .............................................................................. 26

4.5 Distribution and retail stage ........................................................... 27

4.6 Losses in the production chain ........................................................ 27

4.7 Extra emissions due to exclusions ................................................... 27

5 Results .................................................................................................... 28

5.1 Total Carbon Footprint ................................................................... 28

5.2 Results per processing phase .......................................................... 30

6 Emission Reduction Potential .................................................................. 34

6.1 Emission calculations ..................................................................... 34

6.2 Emission reduction within the supply chain ....................................... 34

7 References .............................................................................................. 36

8 Annexes .................................................................................................. 38

Annex 1. Emission factors .................................................................... 38

Annex 2. Soil emissions ....................................................................... 40

Annex 3. Age of farms and representativeness of data ............................ 43

Annex 5. All results ............................................................................. 45


Summary

This study aims to calculate the carbon footprint of bananas originating

from Dole plantations in Costa Rica which are sourced to German

supermarkets.

The approach chosen is “cradle-to–gate”, which means that the raw

materials are the start of the production chain and the study ends at the

retail shelf.

The study describes and defines the sources of greenhouse gas emissions

within the whole supply chain, including the production of input materials

that are used in the different production phases of bananas.

The phases included in the analysis are farming, packaging, inland

transportation, shipping, ripening and distribution.

The methodology that has been used to define the system boundaries and

the allocation method is in line with the PAS 2050, the ISO standard

14044 and verified by TÜV-Nord according to their standard.

Most of the data originates from Dole’s internal systems, including the

data from the overseas transport. However, in some cases estimates have

been made or secondary data has been used.

The results indicate that the carbon footprint of the bananas amounts to

1124 kg CO2e per ton of bananas.

Most of the emissions are related to the overseas transport. Also the

emissions related to fertilizer use have a significant impact on the total

carbon footprint of bananas.


Acronyms and Glossary

Allocation Partitioning the input or output flows of a process or a

product system between the product system under study and one more other product system

Carbon Credit Certificate that represents 1 ton of CO2e that is generated by emission reduction projects

Carbon Footprint Sum of all GHG emissions produced by the product’s life

cycle (within the defined boundary)

Carbon label Quantitative or qualitative label on a product that

displays the carbon footprint of a product

Carbon neutral Emissions related to a product that have been compensated by the purchase of carbon credits

CH4 Methane Gas which has a GWP of 25 CO2e

CO2e Carbon Dioxide Equivalent

Co-products Products that originate from the same raw material

Functional Unit Quantified performance of a product system for use as a reference unit

GHG Greenhouse Gas

GWP Global Warming Potential

Input Products, material or energy flow that enters a unit process

IPPC Intergovernmental panel on climate change

ISO 14044 International standard for life cycle assessments, developed by ISO in 2006

Kyoto Protocol International treaty with the goal of achieving

“stabilization of greenhouse gas concentrations in the

atmosphere at a level that would prevent dangerous anthropogenic interference with the climate system”

Life Cycle Assessment Compilation and evaluation of the inputs, outputs and

the potential environmental impacts of a product system

throughout its life cycle

Life cycle Consecutive and interlinked stages of a product system, from raw materials acquisition to final disposal

N2O Nitrous Oxide which has a GWP of 298 CO2e

http://en.wikipedia.org/wiki/Treaty


Output Product, material or energy flow that leaves a unit process

PAS 2050 First international standard for carbon footprint assessments, developed by Carbon Trust in 2008

Product system Collection of all processes which model the life cycle of a

product

Scope 1 Direct GHG emissions that occur on the company itself

Scope 2 Indirect emissions that occur elsewhere due to the

generation of purchased electricity consumed by the company

Scope 3 All other indirect emissions (e.g. due to production of

purchased inputs and transport) that are related to the

product

System boundary Set of criteria specifying which unit processes are part of a product system

UNFCCC United Nations Framework Convention on Climate Change

VER Voluntary Emission Reduction or Verified Emission Reduction, also called Carbon Credit


1 Public Awareness and Market Developments

Over the past few years, interest in global warming and climate change has

grown exponentially. People have realized that it is time to act. The Kyoto

Protocol1 - initially adopted for use on 11 December 1997 in Kyoto, Japan and

entering into force on 16 February 2005 – requires specific countries (developed

nations) to meet reduction targets of GHG emissions relative to their 1990 levels

during the period of 2008 to 2012. Thus, many governments are developing

initiatives to reduce GHG emissions through national policies, such as emissions

trading programs, carbon or energy taxes, regulations and standards on energy

efficiency and emissions. Global warming and climate change are viewed as part

of any sustainable development strategy. It is therefore crucial for companies to

be able to identify, analyze and understand their GHG risks in order to develop a

sustainable risk management strategy. Only then can companies be successful in

a competitive business environment for the long run.

Besides politicians, governmental authorities and organizations, another powerful

group - which is the “consumer on the street” - increasingly pays attention to

issues related to climate change and sustainability. The consumers’ expectations

have changed. Consumers will give preference to products which offer more than

just a single quality claim. Recently published studies around the globe clearly

show the consumers changing behavior and its willingness to pay more for

sustainable foods. In a study on the “new” lifestyle of health and sustainability

(LOHAS) carried out by Ernst & Young, the authors conclude that “consumers are

not only more demanding about their product choices for their own benefit, but

also realize the direct and indirect impact of their consumption on the

environment. Attributes like “organic farming”, “fair trade”, and “sustainability”

are becoming increasingly anchored in the consumers mind (LOHAS, 2004).

To meet these changing consumer expectations and to help mitigate climate

change, many initiatives related to carbon footprinting and carbon labeling were,

and are still being developed worldwide. Involved stakeholders are mainly retail

chains and labeling organizations that are driven by global warming and climate

change on the one hand, and by above mentioned consumer expectations on the

other hand. Nearly one out of two consumers are willing to pay more for

sustainable products (GfK, 2008), and carbon labeling is part of sustainability in

this regard.

The initiator of carbon footprinting and labeling schemes related to climate

friendliness of products was the British retail chain Tesco. Meanwhile, a number

of other initiatives have been developed (figure 1).

In reaction to this, Soil & More International developed a climate neutral product

label for products of which the GHG emissions related to their life cycle were

determined, calculated and offset (figure 2). The calculated carbon footprint of

such products was certified by TÜV-Nord CERT GmbH and launched at Deko and

BioMarché in the Benelux countries in 2007. All these different initiatives clearly

indicate that carbon labeling already is a unique selling point, which buyers and

consumers will acknowledge when making their purchase decision.

1 The Kyoto Protocol is a protocol to the United Nations Framework Convention on Climate Change (UNFCCC or FCCC), an international environmental treaty with the goal of achieving “stabilization of greenhouse gas concentrations in the atmosphere at a level that would prevent dangerous anthropogenic interference with the climate system” (www.unfccc.com).

http://en.wikipedia.org/wiki/Kyoto

http://en.wikipedia.org/wiki/Japan

http://en.wikipedia.org/wiki/Protocol_(diplomacy)

http://en.wikipedia.org/wiki/United_Nations_Framework_Convention_on_Climate_Change

http://en.wikipedia.org/wiki/Environmental_policy

http://en.wikipedia.org/wiki/Treaty


A list of worldwide initiatives related to carbon footprinting and carbon labeling is

displayed below (www.climatechangecorp.com):

BSI Standards Solutions led development of PAS 2050 (Publicly Available

Specification) at the request of Defra (Department for Environment, Food

and Rural Affairs) and the Carbon Trust. PAS 2050 is a standard for carbon

footprint assessment of goods and services and to date, it is the most

comprehensive standard that provides guidelines on how to assess carbon

footprint of a product and services. Tesco, the British retail chain, was the

largest test of the PAS 2050 draft product carbon footprinting method and

the Carbon Trust Carbon Reduction Label.

The French supermarket chain Casino was the first supermarket in Europe

that initiated carbon labeling. The label is called “l’Indice Carbone” and is a

quantitative label for their own products. The methodology was developed

by the Bio Intelligence Service (Bio IS). The supermarket chain E. Leclerc

was also one the first French companies to introduce voluntary carbon

labeling. Both these French labels were developed in cooperation with the

French Environment and Energy Agency (Ademe).

Migros, a Swiss supermarket chain, introduced its own product labeling

program. Migros label “Climatop” is displayed on the supermarket’s

products if the product’s emissions are 20% lower than those of its

counterparts within the same product category.

The European Eco label also aims to indicate the amount of emissions on

its label. However, the label will not be quantitative but qualitative.

In the US, Carbon Fund created a label called “Certified Carbon Free”. This

was done in cooperation with Carbon Trust and the calculated carbon

footprint of the product was offset.

The Californian Climate Conservancy, a spin off from Stanford University,

developed the “Climate Conscious” label, rating products and the GHG

emissions due to their production, thus classifying products into gold,

silver or bronze products.

KRAV (organic standards association) in Sweden is currently developing a

carbon footprint label for locally produced food products.

In Germany, WWF (World Wide Fund for Nature), the Öko-Institut, PIK

(the Potsdam Institute for Climate Impact Research) and THEMA 1

coordinated a pilot project for a labeling scheme for carbon labels.

The Californian Carbon Label aims to standardize the carbon footprint of

companies and oblige companies to give their input and output data.

The Canadian organization CarbonCounted developed a Carbon Counted

logo for companies.

Other countries that are assessing carbon labeling are China, Japan,

South-Korea, Australia and Finland.

Figure 1. Different carbon labeling initiatives that have been launched recently.

http://www.climatechangecorp.com/


The Soil & More label differentiates from other labels, as the offsetting takes place

within the agricultural sector. Soil & More’s emission reduction projects are

composting projects in developing countries, where green waste is composted to

ensure soil fertility. In Soil & More’s vision, soil degradation is one of the world’s

most important environmental and agricultural problems and composting of

organic materials enhances soil fertility, reduces climate change effect, water use

and amount of waste.

Figure 2. If the CO2 emissions which occur within the supply chain of a product are compensated by the purchase of an equal amount of carbon credits (verified by TÜV-Nord), products can be labeled with the Climate Neutral logo displayed above.


2 General information

2.1 Introduction

This carbon footprint calculation was carried out by Soil & More International B.V.

and its representative, Boki Luske, upon request of Dole Food Company.

This study aims to layout and calculate a comprehensive CO2e (carbon dioxide

equivalent) footprint of bananas. The term “carbon footprint” stands for the total

sum of all greenhouse gas emissions caused by a product’s life cycle. The system

boundaries of the footprint are defined in section 3.2.

The term greenhouse gas emissions stands for compressible fluids that were

attributed a coefficient for their global warming potential by the

Intergovernmental Panel on Climate Change (IPCC). This study includes different

greenhouse gases that are emitted during different stages of a product’s life

cycle: carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O), CFCs and

HCFCs. In the following footprint, all identified greenhouse gases are converted

into CO2e by multiplying them with the GWP value.

Table 1. Greenhouse gasses included in this study with their GWP value

Type of gas Chemical formula GWP 100

Carbon dioxide CO2 1

Methane CH4 25

Nitrous oxide N2O 298

Hydrochlorofluorocarbons HCFCs 124-14,8002

Chlorofluorcarbons CFCs 4,750-14,400

2.2 Company Profile

Dole is one of the world’s largest fresh food producing and distributing

companies. The company consists of three main businesses: fresh fruits,

packaged foods and fresh vegetables. The headquarters’ office is in California,

USA. In Costa Rica, the company operates through a subsidiary called Standard

Fruit de Costa Rica S.A., which produces and sources mainly bananas and

pineapples (conventional and organic) and employs approximately 7,000 people.

2.3 Background information

This assessment covers the carbon footprint of bananas that are produced on

banana plantations of Dole in Costa Rica and are exported to Germany. For the

assessment it was assumed that the bananas were sold in a supermarket chain in

Germany.

In Costa Rica, Dole has approximately 7,000 hectares of banana plantations. The

company owns six banana farms and sources bananas from approximately 20

independent growers representing an additional 6,500 hectares. Approximately

65% of the bananas are shipped to the US, with most of the remaining exported

to the European market.

The bananas are harvested when they are green. After washing they are packed

in carton boxes and transported in containers to the port for export.

2 Depends on the specific type of gas.


After arrival in the terminal in Puerto Moín, the refrigeration units on the

containers are connected to an electric power source. Refrigeration delays the

banana ripening process. When the European vessel arrives in the port, it is

loaded with bananas that have been stored at the terminal yard and also by

containers that arrive directly from the farm to the vessel. Virtually all the

bananas travelling to Europe are loaded on pallets into the refrigerated holds of

the ship. After arrival in Europe (Antwerp or Hamburg) the bananas are

discharged and stored in refrigerated warehouses. They are subsequently

dispatched to different ripening facilities in Hamburg and Roisdorf. After ripening,

the bananas are then transported to several distribution centers in Germany and

then on to supermarkets.

2.4 Goals of a carbon footprint

This assessment results in the carbon footprint for bananas. The goal is to

identify sources of greenhouse gas emissions and to calculate the amount of such

gases emitted due to the assessed banana life cycle. The carbon footprint serves

to identify the environmental performance of a specific product as to greenhouse

gas emissions, thus assessing its impact on climate change.

Further goals of this carbon footprint are:

to collect information for the company in order to reduce GHG emissions

to identify cost saving opportunities for the company

to demonstrate environmental and responsible leadership of the company

to receive recognition for an early voluntary action

to respond to changing consumer expectations (see chapter 1)

The comprehensive carbon footprint has been verified and certified by an

independent third party, TÜV-NORD CERT GmbH. TÜV-NORD is accredited by the

UNFCCC (United Nations Framework Convention on Climate Change)3 to carry out

audits of emission reduction projects and therefore has the technical knowledge

on climate issues to carry out certifications.

Upon certification of the assessment by an accredited third party, the emissions

related to the product’s life cycle may be offset through the purchase of emission

rights or carbon credits.

2.5 Functional Unit

A functional unit is the quantified performance of a product for use as a reference

unit in a given assessment4. For this comprehensive carbon footprint assessment,

the functional unit was identified to be 1 ton of final product. The term “final

product” refers to the product which is placed on the retail shelf, ready to sell.

Therefore, all greenhouse gas emissions caused by the primary production stage,

the processing stage, and the retail and transportation stage of bananas are

broken down to a quantified unit of 1 ton final product.

3 Signed in 1992 at the Rio Earth Summit, the UNFCCC is a milestone Convention on Climate Change

Treaty that provides an overall framework for international efforts to mitigate climate change. The Kyoto Protocol is a protocol of the UNFCCC. 4 Guide to PAS 2050, how to assess the carbon footprint of goods and services, page 57.


3 Methodology

3.1 General methodology

The methodology used for this assessment was developed by Soil & More and is

based upon the ISO 14044 standard and the PAS 2050.

3.2 System boundary and scopes

In this chapter, the system boundary, as well as the scopes of the assessed

bananas will be described. The term boundary refers to the set of criteria

specifying which unit processes are part of a product’s life cycle and are therefore

accounted for in the carbon footprint of a specific product. Once the system

boundary has been defined, the greenhouse gas emissions arising during the

different stages of the product’s life cycle will be identified and assigned to three

different scopes, as introduced by the WRI (World Resource Institute) and the

WBCSD (World Business Council for Sustainable Development) in the Greenhouse

Gas Protocol.

System boundaries

The carbon footprint includes the greenhouse gas emissions that are released

during different stages of the life cycle of bananas. The inputs and outputs are

analyzed for every production stage, and emissions related to production and

transportation are calculated.

The emissions that are directly emitted during one stage as well as the indirect

emissions are taken into account. For instance, the combustion of fossil fuels

causes a direct emission in a production or transport phase but the production of

fossil fuels is also related to greenhouse gas emissions. The latter one is called an

indirect emission.

For every production stage the inputs and outputs are inventoried. This means

that the yield (of main and co-products) is inventoried, just like the amount of

discarded products.

The following stages in the life cycle of bananas are included:

Stage 1: Farming

o Energy consumption: electricity and diesel/petrol for tractors and

other equipment

o Soil emissions (see annex 2 for the calculation method of direct and

indirect soil emissions related to fertilizer use)

o Transport to next stage

Indirect emissions due to the manufacturing and

transportation of agricultural inputs

Indirect emission due to the generation of used energy

Indirect emission due to the production and transport of

used fossil fuels

Stage 2: Packing

o Energy consumption: electricity and diesel/petrol use

Indirect emissions due to the manufacturing and

transportation of packing inputs (e.g. packaging materials)


Indirect emission due to the production and transportation

of used fossil fuels


Stage 3: Export, Ripening and Distribution

o Refrigerants use during transportation

o Energy consumption: electricity and diesel/petrol use


Indirect emission due to the production and transportation

of used fossil fuels

Indirect emission due to the production of ethylene

Figure 3. System boundaries used in this assessment.

Scopes:

In line with the approach Greenhouse Gas Protocol, the emissions identified

within the system boundary and the different stages are assigned to three

different scopes as follows:

Scope 1: Scope 1 emissions include the direct GHG emissions of a

company. These emissions arise from sources that are owned or controlled

by the company.

Scope 2: Scope 2 emissions include indirect GHG emissions of the product.

These are emissions from the generation of purchased electricity

consumed by the company. Purchased electricity is defined as electricity

that is purchased or otherwise brought into the organizational boundary of

the company. Scope 2 emissions physically occur at the facility where

electricity is generated.

Scope 3: Scope 3 emissions include other indirect GHG emissions of the

company. These emissions are consequences of the activities of the

company but occur at sources owned or controlled by another company.

S T A ND A R D


3.3 Data sources

Different forms of data may be taken to carry out a comprehensive carbon

footprint. The most commonly used types of data are:

Primary activity data: data taken from documents or computer systems

(financial accounting) that are directly linked to the specific assessment,

such as electricity invoices to calculate emissions caused by electricity.

Secondary data: such as databases, studies, and reports.

Assumptions: assumptions made based on internationally recognized

standards and studies.

Wherever possible, primary data was used to carry out this carbon footprint. In

case such primary data wasn’t available, secondary data was used. In case the

sources of this secondary data proved to be unreliable, assumptions were made.

The analysis of data was carried out on the basis of the following criteria:

Completeness: a comprehensive carbon footprint assessment must be

based on complete data, as too many assumptions might distort the final

result.

Reliability: data must be taken from reliable sources. It should be

transparent and traceable.

Accuracy: data must be as accurate as possible, including data related to

the specific process, product or company.

Time frame: data must be taken from one particular, clearly defined

period in time (which is usually a period of 12 months).

Geographical affiliation: specific data for the assessed region or country

must be taken for the assessment.

In the case of uncertainty or several different data sources, the most

conservative approach was chosen. That means the value causing the highest

amount of emissions was taken for the calculation.

In general, it is advisable to use as much primary data as possible so that the

actual emissions can be quantified in a more understandable way and

opportunities to improve efficiency can be easily identified.

For this assessment all primary activity data was collected from the year 20085.

Dole has a comprehensive accounting system that registers the costs of all the

products that are used and produced within the company. Apart from this system,

Dole has been working on the determination of their greenhouse gas emissions.

The calculations are connected to the accounting system and the same computer

system has been used to derive the data for this assessment. In some cases,

such as the use of plastic on the farm, not all information was available and

estimations were done.

The analyzed plantations, packing facilities, the terminal in Puerto Moín, the port

and the aerial spraying facilities were visited by the author of this study in 2009.

The secondary data, such as emission factors related to the production of input

materials were derived from different databases and studies. The year, accuracy

and age of this data are also mentioned in the assessment.

5 Unless stated otherwise


3.4 Allocation with co-production

In general, during different processing steps, one raw material is used to produce

many different products (co-products) and the “upstream” emissions (gasses that

are emitted in earlier processing steps) have to be assigned to the different

products. This division of upstream environmental effects is called “allocation”

and can be done in several ways.

In this study, economic allocation is used for situations where the absolute or

relative price and the mass balance are known. A combination of the two

determines the allocation factors as shown in the table below. Mass allocation is

used in the situation where prices are unknown, or in situations where economic

allocation is not applicable (for instance, transport of different products in one

truck or packaging of many different products in one pack house). In this

document the used allocation method and allocation factors are mentioned for

each process phase.

Mass balance of

co-products and mass allocation factor

Relative economic value of co-products

Mass* price Economic allocation factor

Co-product 1 60% 1 60 43%

Co-product 2 40% 2 80 57%

Total 100% 140 100%

3.5 Exclusions

In general the following sources of emissions are not included in the carbon

footprint:

The user phase of products is not included in the carbon footprint, as it

has a high level of uncertainty and the estimated emissions during the

user phase on the eventual carbon footprint is only minor.

The waste/recycling phase of the product is excluded because of high

uncertainty.

Land use change is a major worldwide source of greenhouse gas

emissions, but the methodology on how to allocate it to different products

is still under development and is therefore excluded in this assessment6.

Emissions from the production of capital goods (such as trucks, airplanes

and buildings).

Travelling of employees to and from their place of work.

Human energy requirements.

Animals providing transport services.

Transport of consumers to and from retail.

6 According to the PAS2050, land use changes from less than 20 years are taken in to account. The

last banana plantation developed by Dole was Bananito farm in 1990. The banana plantations La Estrella and Rio Frio date back from 1955 and 1968 (annex 3).


4 Greenhouse Gas Inventory

4.1 Farming stage

Location of the banana plantations

The banana plantations of Dole in Costa Rica are all situated on the Eastern side

of the country near the Atlantic Coast. Bananas that are of export quality are

packed at the plantations and transported by rail or by trucks to Puerto Moin.

From here the product is shipped to different continents. Some of the yield is not

suitable for export and is sold to a local company that makes processed banana

products.

For this study, data from two banana plantations was inventoried and used to

calculate the average. Valle de la Estrella and Rio Frio are banana plantations of

Dole covering 1,384 and 1,479 hectares respectively. The plantations are situated

in the south eastern and central eastern region of Costa Rica.

Figure 4. Geographical location of the banana plantations in Costa Rica.

Process description

The banana plantations of Dole are monocultures of the Cavendish species.

Because the plants are triploid, they don’t produce seeds but are propagated from

offshoots. After the first planting of banana shoots, it takes 9 months until the

first bunch of bananas can be harvested. When a mother plant is harvested the

plant is cut and the offshoot or “sucker” that is already present grows next to it

and starts to grow and develop fruit. The offshoot can be harvested 14 weeks

later. This system can go on for many years. At a banana plantation only one

sucker is allowed to grow at one time from a mother plant, so if more suckers

develop they are removed. The banana bunches are harvested by hand. The

banana stems are attached to steel cables and special pulleys which are hauled

by mules from the plantation towards the packing facility that is located on the

farm itself.


Figure 5. Picture of a banana plantation where a mother plant and a sucker are visible (left photo). At the packing facility banana bunches are quality checked, washed, cut into consumer units and packed (right photo).

Yields

In 2008 Valle de la Estrella and Rio Frio together yielded in average 45.78 tons

per hectare. Furthermore, a portion of the harvested bananas didn’t meet the

export quality specifications and were sold to a company specialized in the

production and distribution of processed banana products. The economic

allocation factor was calculated based on volume and relative prices of export

quality bananas and the one sold to an independent company (table 2).

Table 2. Average yields per hectare and allocation factors of the plantations Valle de la Estrella and Rio Frio in the year 2008.

Yield (ton/ha) Price ratio Allocation factor

Exported bananas 45.78 37.48 99.7%

Bananas sold to an independent company

5.39 1 0.3%

Total 51.17 100%

According to the FAO, the average banana yield in Costa Rica amounted to 46.3

tons/ha in 2006 and 47.5 tons/ha in 2007 (www.faostat.fao.org).

Nutrient requirements

Bananas have high requirements for both nitrogen and potassium. According to

literature (Soh, 1997) bananas have the highest fertilization rate per hectare of

all food crops. Most of the fertilizers at the banana plantations of Dole are applied

by hand. Just a small amount of the fertilizers are foliar fertilizers which are

applied by aerial sprayings by the company AFCA (Aerofumigación

Centroamericana S.A., see below).

Most of the Nitrogen (N) applied is in the form of urea (57%) and the rest as

ammonium nitrates. These fertilizers are important because they are related to

nitrous oxide emissions from the soil and the production of chemical nitrogen

fertilizers which are very energy intensive. In relation to emissions of greenhouse

gasses, the application of calcium carbonates is also important because the

carbonates react in the soil and causes CO2 emissions (IPCC, 2006).


Pest and disease control

Especially in humid countries such as Costa Rica, banana plantations use

relatively high amounts of fungicides to protect the plants for the Black Sigatoka

fungus which causes significant loss of leaf area and yield reductions. The

company that performs the sprayings for Dole is called AFCA. AFCA uses a GPS

system to optimize the spraying activities that automatically open and close the

spray valves of the plane during flight. On average, the applications are done

every 4 to 7 days depending upon the incidence of the disease.

Jet fuel use

The jet fuel used for aerial sprayings at Estrella in 2008 amounted to 188,716.8

liters; this means a use of 136.4 liters/ha. In the case of Rio Frio 170,040.9 liters

of jet fuel was used, which means a usage of 115.0 liters/ha. The average jet fuel

use is displayed in table 6. According to literature, a typical spray plane uses 20-

45 gallons of aviation gasoline per hour spraying 60 hectares per hour. That

means that on average 1.9 x 60 = 113.6 liters/ha of aviation gasoline is used to

spray one hectare during one growing season. So, it can be concluded that the

primary activity data is in the same order of magnitude as the literature suggests.

Figure 6. Planes for aerial sprayings run on jet fuel.

Diesel and gasoline use

Since aerial applications cannot cover the sides of the banana plantations, the

sprayings for these zones are done using ground spraying equipment. The fuel

used at Estrella in 2008 amounts to 1,167 liters of diesel and 22,440 liters of

gasoline. For Rio Frio fuel use amounted to 20,117 liters of diesel in 2008. No

gasoline was used at Rio Frio. For commuting to and from the plantation, the two

farms used 62,859 and 8,371 liters of gasoline respectively. The 2008 average

fuel use per hectare can be found below.

l/ha

Jet fuel 125.66

Diesel 7.22

Gasoline 8.11

Gasoline for commuting 25.54

Electricity use

Electricity use on the farm is mostly used in the packing facility and is therefore

displayed in section 4.2.


Use of plastics

To protect the banana bunches from insects during their growth they are covered

with plastic. To minimize plastic use, the plantations in Costa Rica re-use the

plastic. On average, one bag can be used 3 times before it is damaged and

needs to be renewed. In case of damage the plastics are collected and recycled

by the company Recyplast. The use of plastic per hectare has been estimated by

multiplying the number of bunches that are harvested per hectare per year with

the weight of a plastic bag (0.025 kg). Then a re-use rate of 30% was assumed.

The yearly amount of plastic was therefore estimated as 17.77 kg LLDPE (linear

low-density polyethylene) per ha/yr.

Transport to the packaging facility

Transport by mules of the banana bunches from the field to the packing facility

occurs via steel cables that move like a conveyer belt. As mentioned in the

methodology section, emissions due to animals are excluded from the

assessment. The packing facility is located on the farm itself. To protect the

bananas from damage, plastic is put in-between the banana clusters. This plastic

is recycled at the plantation.


4.2 Packing stage

Process description

The packing facilities are located at the banana plantations. Before entering the

packaging process, the banana bunches are quality checked. Banana bunches can

be rejected if severe insect, bruising or maturity issues are observed. An

evaluation of internal ripening is performed by cutting a finger from the second

hand of all stems from the oldest age being harvested. If premature ripening is

identified by observing the color or softness of the flesh, the stem is rejected and

not packed. The banana bunches that pass this initial quality check are cut into

hands consisting of 5-7 fingers (table 4 on the next page). After that, the

bananas are washed in water basins with a minor percentage of chlorine. In many

of the packing plants, the water is recycled and reused up to six packing days

within the facility. The banana hands can be individually wrapped in plastic bags

(these are called “consumer bags”) or put together in a large bag and packed in

corrugated cardboard boxes. The boxes are packed together on pallets; the

standard pallet contains 48 boxes. 20 pallets are then loaded into each container

that is transported from the farm to the terminal in Moín, situated on the Atlantic

Coast of Costa Rica (table 5 on the next page).

Figure 7. Packing facility at Valle la Estrella (rinsing, cutting, packing, palletizing).


Table 4. Terminology regarding to banana packing.

finger cluster hand bunch

Banana (finger) 1 5-7 15-20 200

cluster 1 3 30

hand 1 8-14

Table 5. Weight per different units of bananas.

box pallet container

kg 18.14 870.72 17.414

box 1 48 960

pallet 1 20

Electricity use

In the packing facilities the bananas are transported on conveyer belts powered

by electricity. The main use of electricity is for lighting and for operating the

pumps that extract water from the wells. The electricity is taken from the national

grid.

Fuel use

Sometimes the power supply fails and power is generated on the farm itself by

using diesel generators. Table 6 displays the average fuel use for electricity

generators for the packaging facilities of Valle la Estrella and Rio Frio in the year

2008.

Table 6. Average electricity use and fuel use per ton of product in the year 2008.

kWh electricity from grid L diesel

input/ton banana 13.93 0.11

Packaging material

The corrugated boxes weigh 1.286 kg and the plastic 0.025 kg, for each unit of

18.14 kg of bananas. The boxes and plastic used to be produced from paper and

plastic material in a facility that was owned by Dole in 2008, but the facility was

recently sold to a third-party. The facility uses paper, plastic, electricity and

bunker fuel to produce the boxes. For this assessment default emission factors for

the production of corrugated board and plastic were sourced from an online LCA

database (see annex 1).

Table 7. Use of packaging materials per ton of bananas packed.

Corrugated board Plastic (LLDPE) kg material/ton banana 70.9 1.4

Transport to the terminal

After packing, the bananas are transported to the container terminal yard in

Moin. There are two means of transport that are used. Most of the containers with

bananas are transported by trucks (67%). However, some of the farms have a

rail line available next to the packing facility. The railroad, owned by the National

Railroad Company (Incofer) is directly connected with the terminal yard. Although

the trains move slower than trucks and use diesel, this form of transportation is

more energy efficient than road transportation. Fuel use during transport in Costa

Rica is displayed in annex 3. The transport of bananas in Costa Rica does not use

refrigeration, as green bananas can easily resist going without cold storage for

over 24 hours.


4.3 Terminal and port operations and overseas transport

Process description

After transportation by train and truck, the containers with bananas arrive at the

Dole terminal, located in Moin and very close to the port. The containers are

rinsed, taken off the train or trucks and connected to the grid for refrigeration.

Refrigeration at 57˚ Fahrenheit (14˚ Celsius) prevents the ripening process of

bananas. The container reefer units use refrigerants in their cooling systems.

After a storage ranging from hours to several days, the containers are

transported to the dock. Sometimes the containers are directly loaded on the

ships without connection to the grid, particularly in those cases where product is

shipped to the European Union. Bananas shipped to the US are mostly shipped in

containers, but in the case of transportation to Europe, pallets are taken out of

the containers and separately loaded below decks into reefer vessels. For Europe,

Dole uses two shipping services: Hamburg and Antwerp.

Electricity use

The electricity use in the terminal yard amounted to 21,695,569 kWh in 2008.

With a turnover of 63,075 containers that were connected to the grid in the

terminal, the average electricity use per container is approximately 344 kWh.

Table 8 displays the electricity use per ton of bananas.

Fuel use

Transportation at the terminal and in the port requires fuel use. In 2008 the total

fuel use amounted to 652.457,28 liters of diesel. Table 8 also displays the

average diesel use per ton of bananas.

Table 8. Electricity and fuel use during terminal and port operations in 2008.

l diesel kWh input/ton banana 0.59 19.75

Refrigerant use

With the data available it was not possible to identify the exact amount of

refrigerant that is used for storage of bananas in the container terminal. The

reason for this is that bananas are stored for varying time periods at the terminal

and the container units travel back and forth from the tropics to the market

countries (especially the U.S.). Much of the refrigerant use is due to the

transportation of pineapples which do require low temperature conditions after

being packed. Unlike bananas, pineapples are harvested ripe and therefore need

to be refrigerated during transport. With the data that was available, a

conservative estimate has been made on refrigerant use amounting to 12.22 kg

CO2e/ton banana.

Figure 8. Storage of containers at the terminal and port operations.


Shipping routes

Approximately 50% of the bananas that are shipped to Germany are exported

from Puerto Moin in Costa Rica to Hamburg. On the Hamburg service, the vessels

make a stopover in Lisbon, where 10-25% of the fruit is unloaded. This stopover

is also made during the return trip to the Americas. Furthermore, the vessels

make a stopover in the port of Turbo, Colombia, before returning to Puerto Moin

(see figure 9, left hand side). The other 50% of the bananas which supply

Germany are shipped to the port of Antwerp. Dole’s Antwerp service leaves from

Puerto Moin and has a stopover in San Juan, Puerto Rico. This stopover is also

made during the return trip to the Americas. After San Juan, the vessels visit the

port of Castilla (Honduras), and then return to Puerto Moin (see figure 9 right

hand side).

Figure 9. Dole’s Hamburg (left) and Antwerp (right) services visit different ports.

Data used

For the calculation of the shipping emissions, primary activity data on fuel use

and cargo quantities were used from Dole for the years 2007 and 2008. The data

originates from the Integrated Maritime Operations System (IMOS), which is an

elaborate accounting and vessel tracking system used by Dole. Recently the data

from this system has been extracted to calculate the greenhouse gas emissions of

overseas transport for Dole.

Regarding fuel use, actual routes and the amount of cargo on the ships, the data

originating from Dole-owned ships is more accurate than the data coming from

third parties. The third-party emissions have been computed using a model (IMO,

2008) and assumptions were made to enable these emission calculations. During

shipping by third party ships, Dole fruit is just a minor part of the total cargo and

the amount of other cargo was often unknown. Because it was estimated that

shipping has a significant impact on the total greenhouse gas emissions of

bananas, it was decided that the most accurate data (primary activity data) was

preferred over secondary data. Therefore it was decided to use the primary

activity data, originating from the IMOS system used by Dole.

The percentage of fruit shipped by Dole and by third parties differs over the

years. In the case of bananas, 98% was shipped from Puerto Moin to Hamburg by

third parties in 2008. In 2007, it was the opposite: 99% was shipped with Dole-

owned ships (table 9 on the next page). This means that the data from the IMOS

system of the year 2008 for shipments to Hamburg was not representative. It

was therefore decided to use the weighted average of the shipment data to

Hamburg for the year 2007, which is based on 50 shipments.


In the case of shipments to Antwerp, in 2008 around 90% of the fruit was

shipped with Dole owned vessels. Therefore it was decided to use the weighted

average of the shipment data of 2008, based on 52 shipments.

Table 9. Quantities of Dole fruit shipped from Costa Rica to Hamburg and Antwerp.

Puerto Moin-Hamburg Puerto Moin-Antwerp

Total fruit shipped in 2008 (tons) 102,103 171,134

Dole owned vessels 2% 89%

Third party vessels 98% 11%

Total fruit shipped in 2007 (tons) 110,476 183,602

Dole owned vessels 99% 91%

Third party vessels 1% 9%

Fuel use during shipping and hotelling of vessels

The IMOS system differentiates the amount of fuel used during shipping and port

calls (table 10). For the different years and the different services there were

some deviations in these figures. Therefore it was decided to calculate the

weighted average of the bananas that were shipped to Germany (table 11).

Table 10. Average fuel use and cargo load on Dole’s Hamburg and Antwerp services. Puerto Moin-Hamburg Puerto Moin-Antwerp Average

Year 2007 2008

Heavy fuel oil use (l) 50,649.282.76 62,109,319.94 56,379,301,35

Yearly turnover (tons) 233,170.58 345,783.04 289,476,81

Heavy fuel oil (l/ton) 217.22 179.62 194.76

Cargo load during shipping

During most Dole shipments, cargo for internal operations is also transported (so

called ‘proprietary cargo’), just like cargo from third parties (‘commercial cargo’).

Dole has determined the amount of pallet space that is used for different cargo

types, expressed in pallet equivalents. This data allowed for calculation of the

weighted average of the fuel use per loaded pallet equivalent per shipment. The

fuel use was only allocated to the loaded pallets. Emissions for “transporting”

empty space on the vessels were therefore evenly divided over the loaded pallets,

not taking into account the type of cargo.

Allocation during shipping

In the case of bananas, one pallet consists of 48 boxes, each containing 18.14 kg

of bananas. For the calculation it was assumed that all loaded pallet equivalents

were used for bananas and loaded with 0.87 ton bananas. Although in reality

this was not the case, it was found to be the most straightforward and

conservative way of allocation7. If the emissions were evenly distributed over all

pallets, including the empty cargo spaces, the emissions per ton banana

transported would be approximately 35% lower.

Refrigerant use during shipping

7 In Dole’s emission accounting system, another allocation method has been used which allocates

emissions from shipping to fruit, commercial cargo, proprietary cargo and the ‘ocean carrier’. The

latter means the empty space during shipping. The emissions allocated to the ocean carrier are often a significant part. The way the emissions are allocated in this study is therefore more conservative.


Data was available on refrigerant use during shipping from 4 ships that are used

on the European services in the year 2008. The average refrigerant use per ton

banana transported is 8.34 kg CO2e. While these vessels are over 10 years old,

the amount of refrigerant use is rather conservative.

After arrival in Europe

The bananas that arrived in the port of Hamburg are transported to a ripening

facility located within the port (over a distance of 1.5 km). The bananas arrived in

the Port of Antwerp are transported by truck to the ripening facility in Hamburg

(about 20%) or to a ripening facility in Roisdorf/Bornheim (about 30%). 24 tons

of bananas per truck are transported with a fuel use of 0.3 liters of diesel/km.

The trucks return empty to the ports. The weighted average transport distance is

displayed in table 11. Using the weighted average distance and the fuel use, the

fuel use per ton transported is 4.58 liter/ton.

Table 11. Transport distances from port to the ripening facilities.

Distance of return trip (km) % of bananas

Hamburg-Hamburg 3 50%

Antwerp-Roisdorf 476 30%

Hamburg-Roisdorf 1112 20%

Weighted average 366 100%

Figure 10. Transport from the ports to the ripening facilities.

1 transport a week

2 transports a week

3 transports a week


Refrigerant use during transport in Europe

The bananas are refrigerated during transport by truck in Germany. The

refrigerant gas R404A is used which is a colourless odourless mixture of CFCs and

HCFCs (chlorofluorocarbons and hydrochlorofluorocarbons). The Global Warming

Potential (GWP) of these gasses is very high (table 12). For this specific mixture a

GWP of 3,921 was calculated. The yearly use is 2.3 kg of refrigerant can be used

for 1,500 hours of cooling. Per hour cooled transport this means a use of 1.5

grams of refrigerant per truckload.

Table 12. Compounds and GWP of the refrigerant gas R404A (http://cameochemicals.noaa.gov/chemical/26023; PAS 2050).

compound Percentage present in refrigerant gas R404A

GWP for 100 year- time interval

% x GWP

CHF2CF3 (HFC-125) 44% 3,500 1,540

CH3CF3 (HFC-143a) 52% 4,470 2,324.4

CH2FCF3 (HFC-134a) 4% 1,430 57.2

Total 100% 3,921.6

4.4 Ripening stage

Process description

Bananas are imported green into Europe and ripened in specially constructed

rooms by exposure to a controlled atmosphere containing ethylene gas. Ethylene

is a colourless flammable gas with a sweet odour. Ethylene is mostly produce

from naphtha or in a lesser degree from LPG (propane + butane) and NGL

(ethane, propane, butane) (Zimmermann & Walzl) that undergo thermal cracking

with steam. The process is commonly called pyrolysis or steam cracking. The

process takes place within tubular fired heaters that reach high temperatures and

cooling of the product takes place under high pressure. Emission factors for

ethylene production and electricity generation are displayed in annex 1.

Ethylene and energy use

The cooling and heating process requires energy. The quarterly use of ethylene

and electricity was used to calculate the use per ton banana (table 13)8.

Table 13. Ethylene use for the ripening facility.

ethylene use (kg) energy use (kWh) turnover (tons)

per quarter 1,965 722,997.25 5,299

per ton/banana 0.37 136.44 1

8 The optimal energy use for ripening facility in Hamburg was calculated as 1.2 kWh per box. The

amount used in this assessment is based on the measured electricity use and is more conservative.

http://cameochemicals.noaa.gov/chemical/26023


4.5 Distribution and retail stage

After ripening, the bananas are transported to distribution centers of German

retail chains. For this assessment it was assumed that a German retailer

distributes bananas to 35 logistic locations. It was assumed that the bananas

from the ripening facility in Hamburg (70% of the Dole bananas) were

transported to 18 logistic locations that supply the shops in the Northern part of

Germany. Bananas ripened in Roisdorf were assumed to be transported to 17

logistic locations in the Southern part of Germany. The trucks are loaded with 24

tons and are empty when they return. Fuel use for transportation to the logistic

locations was calculated at 5.10 liters per ton banana. Every logistic location was

assumed to supply an average of 270 shops over a distance of 100 km (one

way). In this case the truckload was assumed to be 15 tons and empty on the

way back.

These numbers have been used to estimate the average distance over which the

bananas were transported (table 14). Fuel use for transportation to the retail

shops was calculated at 4.00 liters per ton banana.

Table 14. Weighted average transport distances for distribution of bananas in Germany.

Distance of return trip (km) % of bananas

Ripening facilities-logistic locations

408 100%

Logistic locations-shops 200 100%

4.6 Losses in the production chain

Not every banana produced by the farm is eventually consumed. In all stages of

the supply chain, products can be damaged, lost and discarded. This means that

for the consumption of 1 banana in Europe more bananas have been produced.

For this assessment the average losses in the supply chain were inventoried and

this resulted in the figures presented in table 15.

During data collection in Costa Rica it became clear that quality control at the

packing facilities is very strict, which results in the fact that there is hardly any

loss of bananas between the farm and the port. After ripening the loss is 0.52%.

While information was lacking about losses during distribution, a conservative

assumption was made that 1% of the bananas was rejected during distribution.

Table 15. Losses of damaged products in the banana production chain.

% Discarded Produced for 100% at

retail shelf

Comments

Farming 0% 101.52% Lower quality bananas are included in the

economic allocation factor

Packing 0% 101.52% Lower quality bananas are included in the

economic allocation factor

Storage 0% 101.52% Due to strict quality control at packing plant

Shipping 0% 101.52% Personal communication Theresa Seewald,

Dole Germany

Ripening 0.52% 101.52% Personal communication Theresa Seewald,

Dole Germany

Distribution 1% 101% Conservative estimate

Retail - 100% This assessment does not include unsold

bananas in the shop

4.7 Extra emissions due to exclusions

According to the TÜV-Nord Standard, the results of the carbon footprint had to be

raised by 5% in order to cover excluded emission sources and in order to be

conservative.


5 Results

5.1 Total Carbon Footprint

The weighted average carbon footprint of the bananas which are produced in

Costa Rica and exported to Germany is 1,124 kg CO2e/ton of bananas. When

looking at figure 11 and table 16, the following conclusions can be drawn:

The center of gravity of the emissions is during the overseas transport,

responsible for 62% of the greenhouse gas emissions within the supply

chain.

Second in importance are the emissions related to farming practices which

contribute significantly to the total carbon footprint. The emissions relating

to farm practices are responsible for 12% of the total carbon footprint.

Packaging and ripening are in third place and are both responsible for 8%

of the total footprint.

Minor impacts on the total footprint include transport from ripening to

retail (2%), transport from the farm to the terminal (1%), and terminal

and port operations (2%).

Table 16. Emissions related to the different production phases of bananas.

Phase Kg CO2e/ton of bananas Contribution

Farm 137.83 12%

Packing 89.60 8%

Transport from packing facility to terminal 14.11 1%

Terminal and port operations 25.75 2%

Overseas transport 691.74 62%

Ripening 84.46 8%

Transport from ripening facility to retail 26.62 2%

Extra due to exclusions 53.51 5%

Total 1,123.62 100%

Figure 11. Greenhouse gas emissions per ton of bananas (kg CO2e/ton) within the different phases of the banana supply chain.


The total emissions can also be displayed over 3 different scopes. Scope

1 emissions are directly related to the banana supply chain. Scope 2

emissions are related with the generation of electricity. These emissions

have been emitted elsewhere. Scope 3 emissions are indirectly related to

the banana production chain, for instance emissions due to the

production of fertilizers, corrugated board and ethylene.

When studying figure 12 and table 17 the following conclusions can be drawn:

The major part of the emissions within the banana chain are related to

Scope 1 which include the emissions due to the combustion of fossil fuels

within the supply chain and soil emissions related to fertilization at the

farm level (69%).

18% of the total footprint is indirectly related to the banana chain.

Emissions related to electricity generation, scope 2, and amount to 8% of

the total emissions. One of the reasons that this is just a minor part of the

total footprint is that the Costa Rican power supply has a very low

emission factor.

Table 17. Emissions related to the different scopes of the banana supply chain.

Scope Explanation kg CO2e/ton of bananas Contribution

Scope 1 Burning of fossil fuels and soil emissions related to fertilizer use and fugitive emissions

775.79 69%

Scope 2 Energy generation 87.80 8%

Scope 3 Indirect emissions related to production of input materials

206.53 18%

Extra due to exclusions 53.51 5%

Total 1,123.62 100%

Figure 12. Greenhouse gas emissions per ton of bananas (kg CO2e/ton) divided over the different scopes of the banana supply chain.


5.2 Results per processing phase

5.2.1 Farming and local transport

The total emissions on the farm level amount to 137.83 kg CO2e/ton of bananas

at the retail shelf. When looking at figure 13, the following conclusions can be

drawn:

Within the farming phase most emissions are related to fertilization.

Soil emissions contribute 47% of the footprint on farm level, whereas the

production of fertilizers causes 36% of the emissions (indirectly).

The production of fungicides is responsible for 9% of the emissions on

farm level (indirectly).

Fuel combustion, occurring mostly for aerial spraying, is responsible for

7% of the emissions.

Other indirect emissions related to the production and transport of input

materials have a minor impact on the emissions at the farm level.

Figure 13. Emission sources related to the farming stage of bananas.


5.2.2 Packaging and transport to terminal

The total emissions at the packing facility amount to 103.71 kg CO2e/ton of

bananas at the retail shelf. When looking at figure 14 the following conclusions

can be drawn:

Indirect emissions related to the production of corrugated board boxes are

the major contributor of the total emissions for the packing phase (84%).

Furthermore, transport from the packing facility to the terminal is

responsible for 13% of the emissions.

A minor impact on the total emissions is the generation of electricity.

Figure 14. Emission sources related to the packaging phase of bananas.


5.2.3 Storage, shipping and transport to ripening facility

The total emissions due to storage, shipping and transport to the ripening

facilities amounts to 717.5 kg CO2e/ton of bananas. As mentioned before, the

overseas transport has the major impact on the total footprint and therefore also

on the storage and shipping phase of the production chain (figure 15).

Figure 15. Emission sources related to the storage and overseas transport of bananas.


5.2.4 Ripening and distribution

Emissions related to the ripening phase of the bananas, amount to 111.08 kg

CO2e/ton of bananas. When looking at figure 16, the following conclusions can be

drawn:

Energy consumption at the ripening facility is the major part of the

emissions in the ripening and distribution phase (75%).

Transport related emissions for distribution towards retail cause 22% of

the emissions.

The production of ethylene has only a minor contribution (1%) to the

emissions in this production phase, as well as the production of fuels

(2%).

Figure 16. Emission sources related to the ripening and distribution phase of bananas.


6 Emission Reduction Potential

6.1 Emission calculations

This study is part of an elaborate project where Dole calculates the Company’s

carbon footprint. While a corporate carbon footprint focuses on the total

company, a product carbon footprint focuses on the emissions of the whole

supply chain, allocating it to one functional unit.

It has to be mentioned that a carbon footprint is not an absolute number, but is

an estimate, because system boundaries have to be defined which automatically

means a simplification of reality. Apart from the system boundaries, there is the

issue of time effectiveness that should be taken into account during an

assessment. And a third inaccuracy for a system analyses in general is that

supply chains are dependent on market developments, which means they are

always subject to change.

However, the assessment shows clearly where the center of gravity of

greenhouse gas emissions lies within the supply chain of bananas. Once this is

known, Dole could determine where in the production chain emission reductions

could be reached. Furthermore, the Company may decide to compensate

emissions and market the bananas as carbon neutral.

6.2 Emission reduction within the supply chain

The results of this study clearly show that shipping contributes significantly to

greenhouse gas emissions per functional unit. Thus, it is recommended to explore

the options towards emission reduction at this part of the banana supply chain.

This study is specific in that it uses absolute figures on fuel use during shipment,

which usually are hard to reach. During the data collection in Costa Rica, the

differences with emission factors for shipping in databases were discussed, and it

became clear that fruit is often shipped with relatively small sized vessels. The

reason for this is time pressure. Loading and unloading takes longer for large size

container vessels than for small sized reefer vessels and this is a risk for fresh

produce. Another reason for the relatively small vessels is the size of the port in

Puerto Moin, which is also relatively small. Looking at the fuel use per functional

unit, smaller sized vessels are less efficient than large sized vessels. Shipping of

fruit is therefore in many cases less efficient than the overseas transport of

materials or bulk. Default values from emission databases also assume a high

cargo load per shipment. The primary activity data of this study showed however,

that on average 35% of the vessel’s cargo capacity is unused.

The second option for emission reduction is on the farm level and is related to

fertilization, especially for nitrogen fertilization. The production of these fertilizers

causes significant emissions, just like the biochemical reactions that take place in

the soil after fertilization. Alternative fertilization schemes, such as the use of

leguminous species, could be used to decrease the use of chemical fertilizers and

to improve soil fertility.


Apart from emission reduction within the supply chain, Dole is already working on

several emission reduction projects, such as tree planting and reforestation

activities, forest conservation projects, renewal and replacement of old

containers, and the use of less harmful refrigerant types.

Figure 17. A single banana grown in Costa Rica and sold in a German supermarket has a carbon footprint of 135 g CO2e.


7 References

Baldo, G. L., Marion, M., Montani, M., S-O, Ryding, 2008. Study for the EU

Ecolabel Carbon Footprint Measurement Toolkit. Final Activity Report. Studio

LCE, Italy & SEMC, Sweden

Clemens DR, Weise SF, Brown R, Stonehouse DP, Hume DJ, Swanton CJ (1995).

Energy analyses of tillage and herbicide inputs in alternative weed

management systems. Agriculture Ecosystems and Environment (52): 119-

128.

Davis, J. and Haglund, C. 1999. Life Cycle Inventory (LCI) of Fertiliser Production.

Fertiliser Products Used in Sweden and Western Europe. SIK-Report No.

654. Masters Thesis, Chalmers University of Technology.

FRoSTA, 2009. Fallstudie Tagliatelle Wildlachs. Fallstudie erstellt im Rahmen des

PCF Pilot Projektes Deutschland. http://www.pcf-

projekt.de/files/1257258154/pcf_frosta_tagliatelle_update.pdf

Green, MB (1987). Energy in pesticide manufacture, distribution and use. In:

Stout BA, Mudahar MS, editors. Energy in plant nutrition and pest control.

Amsterdam, Elsevier, p 165-177.

Green and McCulloh (1974). Energy considerations in the use of herbicides. UK,

Imperial Chemical Industries.

Guide to PAS 2050, How to assess the carbon footprint of goods and services,

Carbon Trust

Helsel ZR (1992). Energy and alternatives for fertilizer and pesticide use. In:

Fluck RC, editor. Energy in farm production. Amsterdam: Elsevier, p 3-14

IMO, 2008. Updated study on Greenhouse Gas Emissions from ships. Phase 1

report. Report nr. MEPC 58/INF.6.

Intergovernmental Panel on Climate Change (IPCC), 2006 IPCC Guidelines for

National Greenhouse Gas Inventories:

http://www.ipcc-nggip.iges.or.jp/public/2006gl/index.htm

International Energy Agency http://www.iea.org/Textbase/stats/index.asp

ISO , International Standard on Environmental Performance Evaluation, (ISO

14044), International Standard Organization, Geneva.

IPCC Guidelines for National Greenhouse Gas Inventories, 2006. Volume 4.

Agriculture, Forestry and other land use. Chapter 11: N2O emissions from

managed soils and CO2 emissions from lime and urea application.

Lal, R. 2004. Carbon emissions from farm operations. Environment International

30: 981-990.

LOHAS – Lifestyle of Health and Sustainability (2007), Ernst & Young

Ministerio de Ambiente y Energía, 2007. Estrategia Nacional de Cambio Climático.

Inventario e informe de gases con efecto invernadero (GEI). Programa

piloto para empresas y organizaciones. Version 2. Page 8.

PAS 2050:2008, Specification for the assessment of the life cycle greenhouse gas

emissions of goods and services, Carbon Trust

Protocol 9080 Landbouw bodem indirect t. b. v. NIR 2009, 4D: N20 landbouw

bodem: indirecte emissies, 2009. Ministerie van VROM, Den Haag. found on

www.broeikasgassen.nl

Protocol 9081 Landbouw bodem direct t. b. v. NIR 2009, 4D: N20 landbouw

bodem: directe emissies en beweidingsemissies, 2009. Ministerie van

VROM, Den Haag. found on www.broeikasgassen.nl

http://www.ipcc-nggip.iges.or.jp/public/2006gl/index.htm

http://www.iea.org/Textbase/stats/index.asp

http://www.broeikasgassen.nl/


Soh, K.G. 1997. Fertilizer use by crops. IFA Agro-economics Meeting, Beijing,

China. United Nations, 2005. The Millennium Development Goals Report

2005.

US Environmental Protection Agency (Appendix H of the instructions to Form EIA-

1605)

World Business Council for Sustainable Development (WBCSD)/World Resources

Institute (WRI), 2004, The Greenhouse Gas Protocol, Corporate Accounting

and Reporting Standard, revised Edition

Zimmermann, H. & R. Walzl, Linde AG, Hoellriegelskreuth, Federal Republic of

Germany.

http://media.wiley.com/product_data/excerpt/55/35273038/3527303855.p

df

Wood, S. & A. Cowie, 2004. A Review of Greenhouse Gas Emission Factors for

Fertiliser Production, Research and Development Division, State Forests of New

South Wales

Used websites:

www.faostat.fao.org

GfK Consumer Scope, 2008.

http://www.climatechangecorp.com/content.asp?ContentID=5828

www.unfccc.com


http://lca.jrc.ec.europa.eu/

Gemis 4.2 database, 2004

http://media.wiley.com/product_data/excerpt/55/35273038/3527303855.pdf

http://media.wiley.com/product_data/excerpt/55/35273038/3527303855.pdf

http://www.faostat.fao.org/


http://lca.jrc.ec.europa.eu/


8 Annexes

Annex 1. Emission factors

Table 18 displays the emission factors of the input materials that are used on the

farm. For fossil fuels, the emission due to fuel production and transport was also

included. This emission factor was not extracted from a specific source, but was

conservatively estimated being 10% of the combustion emissions.

Apart from the emissions due to fertilizer production, the Tier 1 method in the

good practice guidelines of the IPCC for the calculation of soil emissions was

included. Both direct and indirect emissions of nitrous oxide due to the application

of nitrogen fertilizers were included.

Table 18. Emission factors used for the assessment.

Kg

CO2e/unit

Source Comments on data quality

Urea (kg N) 4,0 Davis and Haglund, 1999

reviewed in Wood and

Cowie, 2004

Although this study is already over

10 years old, it is the most

comprehensive study on fertilizer

production and CO2e emissions. The EF is more conservative than

EFs from other sources, like

Kongshaug (1998) and Kuesters &

Jenssen (1998). The Emission

Factor (EF) is a European Average

of that time period.

Ammonium Nitrate (kg N) 7,0 Davis and Haglund, 1999


Cowie, 2004

Idem as above.

Urea (foliar fertilizer) (kg N) 4,0 Davis and Haglund, 1999


Cowie, 2004

Idem as above.

Ammonium Nitrate (foliar

fertilizer) (kg N)

7,0 Davis and Haglund, 1999


Cowie, 2004

Idem as above.

Fungicides (kg active

ingredient)

8,0 Lal, 2004 The EF for the production of the

specific active ingredients are

unknown. Therefore the most

conservative known EF has been

chosen for the production of

fungicides (Benomyl). The study

has estimated the emission on the basis of several other studies which

are relatively old (Green, 1987;

Green and McCulloch, 1974; Helsel,

1992; Clemens et al., 1995).

Electricity generation in Costa

Rica (kWh)

0,1165 Ministerio de Ambiente y

Energía, 2007

The EF for electricity generation in

Costa Rica is relatively low, because

most of the energy is generated at

hydropower plants.

Electricity generation in

Germany (kWh)

0,605 IFEU, 2007 found in

FRoSTA, 2009

This source uses the emission factor

for the German Grid of 2005. It was

the most conservative and most

updated emission factor that was

found, coming from a secondary

source however.

Diesel combustion (l) 2,681 US Environmental

Protection Agency (EPA)

Diesel production (l) 0,215 - This is calculated by comparing the

net and gross emissions due to

diesel combustion in the Footprint

Expert Model Framework V3.0a of

Carbon Trust (being 8% higher than

the combustion emissions).

Gasoline combustion (l) 2,327 US Environmental



Gasoline production (l) 0,326 - This is calculated by comparing the


gasoline combustion in the Footprint

Expert Model Framework V3.0a of

Carbon Trust (being 14% higher

than the combustion emissions).

Jet Fuel combustion (l) 2,528 US Environmental


Jet fuel production (l) 0,177 - This is calculated by comparing the


kerosene combustion in the

Footprint Expert Model Framework

V3.0a of Carbon Trust (being 7%

higher than the combustion

emissions).

Heavy fuel oil combustion (l) 3,117 US Environmental


Heavy fuel oil production (l) 0,249 - This is calculated by comparing the

combustion emissions and the total

emissions of diesel in the Footprint

Expert Model Framework V3.0a of Carbon Trust (being 8% higher than

the combustion emissions).

Corrugated board (kg) 1,17 http://lca.jrc.ec.europa.eu/

Ethylene production 1,72 Gemis 4.5 database, 2008 Refrigerant production unknown

http://lca.jrc.ec.europa.eu/lcainfohub/datasetList.vm?topCategory=Materials&subCategory=Plastics


Annex 2. Soil emissions

Nitrous oxide is emitted in soils due to microbial processes in the soil:

denitrification and nitrification. It is a combination of aerobic and anaerobic

processes by microorganisms that use nitrogen and organic matter to feed on.

Soil & More uses the calculation methodology of soil emissions, described in

chapter 11 of the IPCC guidelines on a Tier 1 approach (default methodology).

This methodology is specified by combining it with the Tier 2 approach of the

Netherlands, where different emission factors are defined for different soil types.

The IPCC guidelines define direct and indirect emissions. The emission factors are

dependent on the type of fertilizer, soil type, crop residues and the amount of

fertilizer that is used.

The following parameters are needed in order to calculate the soil emissions:

1. Fertilizer type used (nitrate or ammonia type)

2. Application of fertilizer (kg N/ha)

3. Manure application (kg N/ha)

4. Application method (above or below ground)

5. Crop residues that remain in the field (ton/ha and N content)

6. N binding by legumes (kg N/ha)

7. Soil type (mineral or organic)9

9 Soils are organic if they satisfy the requirements 1 and 2, or 1 and 3 below (FAO, 1998):

1. Thickness of 10 cm or more. A horizon less than 20 cm thick must have 12 percent or more organic carbon when

mixed to a depth of 20 cm;

2. If the soil is never saturated with water for more than a few days, and contains more than 20 percent (by weight) organic carbon (about 35 percent organic matter)

3. If the soil is subject to water saturation episodes and has either:

(i) at least 12 percent (by weight) organic carbon (about 20 percent organic matter) if it has no clay; or

(ii) at least 18 percent (by weight) organic carbon (about 30 percent organic matter) if it has 60 percent or

more clay; or (iii) an intermediate, proportional amount of organic carbon for intermediate amounts of clay

(FAO, 1998).


Direct emissions

The following formula is applied to calculate the direct emissions from agricultural

soils:

CO2-eq (kg/ha) =∑ Eij/ha*EFij*44/28*298

Eij=netto amount of N applied by source i on soil type j (therefore

volatilized N is deducted)

EFij=emission factor of source i on soil type j

44/28= conversion factor of N2O-N to N2O

298= GWP value of N2O

Table 19. Emission factors of direct soil emissions (IPCC, 2006; www.broeikasgassen.nl).

EF mineral soil EF organic soil

Application of mineral fertilizer with nitrate (N2O-N/kg N) 1% 2%

Application of mineral fertilizer without nitrate, with ammonium (N2O-N/kg N)

0,5% 1%

Above ground application of manure (N2O-N/kg N) 2% 2%

Below ground application of manure (N2O-N/kg N) 1% 2%

N-binding (N2O-N/kg N) 1% -

Flooded rice fields (N2O-N/kg N) 0,3% 0,3%

Temperate organic and grassland soils (N2O-N/ha) - 8

Tropical organic grassland soils (N2O-N/ha) - 16

Temperate and boreal organic nutrient poor forests (N2O-N/ha) - 0,6

Cattle, poultry and pigs (N2O-N/kg N) 2% 2%

Sheep and other animals (N2O-N/kg N) 1% 1%

Indirect emissions

Indirect soil emissions are emitted due to leaching of nitrogen and deposition of

volatilized ammonia.

The following formula is applied to calculate indirect soil emissions:

CO2-eq (kg/ha)= ∑ Ei/ha*EFi*44/28*298

Ei= amount of N from source N

EFi= emission factor of source N

44/28= conversion factor of N2O-N to N2O

298= GWP value of N2O

Table 20. Emission factors of indirect soil emissions (IPCC, 2006).

Source EF

Volatilisation and re-deposition of NH3-N and NOx-N)

1%

Leaching of N2O-N 0,75%

Fraction that volatilizes of mineral fertilizers 10%

Fraction that volatilizes of organic fertilizers 20%

Fraction that leaches (if rain or irrigation> water holding capacity)

30%

Crop residues (N2O-N/kg N) 1%


Emissions due to lime and dolomite application

CO2 emissions due to lime application are calculated as follows:

Kg CO2-eq/ha=(Mj/ha* EF limestone) + (Mk/ha* EF dolomite) * 22/6

Mj=kg of limestone applied per hectare

Mk= kg of dolomite applied per hectare

22/6 = conversion factor of carbon to CO2

Table 21. Emission factors due to lime and dolomite applications (IPCC, 2006).

EF

Lime 12%

Dolomite 13%

Emissions due to urea application

CO2 emissions due to urea application are calculated as follows:

Kg CO2-eq/ha=(Mu/ha* EF urea) * 22/6

Mu=kg of N in urea applied per hectare

22/6 = conversion factor of carbon to CO2

Table 22. Emission factors due to urea application (IPCC, 2006).

EF

Urea (N) 20%


Annex 3. Age of farms and representativeness of data


Annex 4. Raw data: inputs used per hectare or per ton.

use per hectare or use per ton

Farm Average

Production export quality bananas (ton/ha) 45.78

Production bananas sold to banana processor (ton/ha) 5.39

Production total (ton/ha) 51.17

total (kg N/ha) 376.96

Calcium Carbonate (CaCO3/ha) 155.59

jet fuel (l/ha) 125.66

Diesel (l/ha) 7.22

Gasoline (l/ha) 8.11

Gasoline (l/ha) 25,54

fungicides 73.00

plastic LLDPE (kg/ha) 17.77

Packing facilities Average

electricity use (kWh/ton) 13.93

diesel (l/ton) 0.11

corrugated board (kg/ton) 70.89

plastic (kg/ton) 1.38

diesel (l/ton) 4.80

refrigerants (kg/ton) 0

Terminal and port operations Average

diesel (l/ton) 0.59

electricity use (kWh/ton) 19.75

refrigerants (kg CO2/ton) 12.22

Shipping Antwerp service

Hamburg service

Average

heavy fuel oil (l/ton) 179.62 217.22 198.42


Transport to ripening facilities Average

diesel (l/ton) 4.58


Ripening Average

ethylene (kg/ton) 0.37

electricity (kWh/ton) 136.44

Transportation to logistic locations average

diesel (l/ton) 5.10

Transport to retail average

diesel (l/ton) 4.00


Annex 5. All results

Phase Source of emission kg CO2 e/ton end product at retail shelf

Farm fertilizer production 49.26

soil emissions 64.80

fuel combustion 9.18

fuel production 0.77

electricity generation -

material production 0.91

plant protection production 12.91

refrigerants -

transport to next phase, fuel combustion

-

transport to next phase, fuel production

-

Packing and local transport


fuel production -

electricity generation 1.65


refrigerants -


13.07

transport to next phase, fuel

production

1.05

Terminal and port operations


fuel production 0.13


material production -

refrigerants 21.68


640.50


51.24

Ripening and distribution

fuel combustion -

fuel production -



refrigerants -


24.65


1.97

5% extra due to exclusions 53.51

Total 1,123.62

soil & more international b.v. -...

Documents